Create several graphs with different geoms() in ggplot2
Create a facet graph
Customize your plots
Create a detailed map
The objective of the lab is to examine the association between weekly average COVID-19 death and vaccination rates in ten regions of the US.
1. READ IN THE DATA
library(tidyverse)
Warning: package 'ggplot2' was built under R version 4.4.3
Warning: package 'tibble' was built under R version 4.4.3
Warning: package 'tidyr' was built under R version 4.4.3
Warning: package 'readr' was built under R version 4.4.3
Warning: package 'purrr' was built under R version 4.4.3
── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr 1.1.4 ✔ readr 2.1.6
✔ forcats 1.0.1 ✔ stringr 1.6.0
✔ ggplot2 4.0.1 ✔ tibble 3.3.1
✔ lubridate 1.9.4 ✔ tidyr 1.3.2
✔ purrr 1.2.1
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag() masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
if (!file.exists("covid_processed.csv"))download.file(url ="https://raw.githubusercontent.com/dmcable/BIOSTAT620W26/main/data/covid/covid_processed.csv",destfile ="covid_processed.csv",method ="libcurl",timeout =60 )covid <-read.csv("covid_processed.csv")
2. PREPARE THE DATA
Since states have different population sizes, make sure that you are working with population-normalized rates (e.g. units cases per 100,000 population). Transform the following variables accordingly: cases, hosp, booster, series, deaths.
Compute the mean by state of the normalized variables above. We recommend creating a new data.frame with the state-averaged data. Keep any other relevant variables from the original data (e.g. population, region etc).
2. Use geom_violin to examine the case rates and death rates by region
You saw how to use geom_boxplot in class. Try using geom_violin instead (take a look at the help). (hint: you will need to set the x aesthetic to 1)
Use facets
Fill color by region
Describe what you observe in the graph
covid_state %>%ggplot(aes(x =1, y = cases, fill = region_name)) +geom_violin() +facet_wrap(~region_name, nrow =2) +labs(title ="Average Case Rates, Per Region")
New York Islands and the Pacific regions had states with the lowest minimum average cases (normalized by population), while the South Central, Southeast, Central Plains, Midwest, and Mountain States had higher median average cases overall (and smaller ranges).
covid_state %>%ggplot(aes(x =1, y = deaths, fill = region_name)) +geom_violin() +facet_wrap(~region_name, nrow =2) +labs(title ="Average Death Rates, Per Region")
Similar to the pattern seen among cases, New York Islands and the Pacific regions had states with the lowest minimum average deaths (normalized by population). The South Central and Southeast regions appeared to have states with higher median average deaths.
4. Use geom_point with stat_smooth to examine the association between time and case rates by region
Filter the data over time to two weeks after the first date in the dataset
Color points by region
Fit a function with stat_smooth by region. Does method=lm or the default method work better?
For the default stat_smooth, play around with the span parameter to get a smooth that fits the data better without being too noisy.
Describe what you observe in the graph
two_weeks <-min(covid$date) +weeks(2)covid_scatterplot <- covid %>%filter(date >= two_weeks)covid_scatterplot %>%ggplot(aes(x = date, y = cases_normalized, color = region_name)) +geom_point() +stat_smooth(span =0.3) +labs(title ="Case Rates Over Time (2020-2021), Per Region")
`geom_smooth()` using method = 'loess' and formula = 'y ~ x'
I believe the default method for stat_smooth works better than method=lm, because you are about to better see the upward/downward trends within the given time frame, instead of a single linear line that only moves upward.
Overall, there is a general trend where the cases rise from the start of 2020 until January 2021, and then there is a decrease in cases that reaches a minimum in June 2021, before another rise in cases is seen again. With the method=lm, the ability to see that decrease in cases in 2021 is lost.
5. Use geom_bar to create barplots of the states by population category colored by region (DESCRIBE!!!!!)
Bars by population category using position="dodge"
Change colors from the default. Color by region using scale_fill_brewer see this
Create nice labels on the axes and add a title
Describe what you observe in the graph
covid_state$pop_high_low <-factor(covid_state$pop_high_low, levels =c("Low Population", "High Population"))covid_state %>%ggplot(aes(x = pop_high_low, fill = region_name)) +geom_bar(position ="dodge") +scale_fill_brewer(name ="Region Name", palette ="Paired") +labs(title ="Number of States with Low or High Populations, Per Region",x ="Low or High Population", y ="Count")
6. Use stat_summary to examine mean vaccination rate and death rate by region with standard deviation error bars (adding another layer???)
Use fun.data="mean_sdl" in stat_summary. Look up what mean_sdl does.
Add another layer of stats_summary but change the geom to "errorbar" (see the help).
Describe the graph and what you observe
Vaccination rates are …
Death rates are…
covid_state %>%ggplot() +stat_summary(aes(x = region_name, y = series),fun.data ="mean_sdl",geom ="errorbar") +labs(x ="Region Name", y ="Mean Series Rate",title ="Mean Vaccination Rate, Per Region") +theme(axis.text.x =element_text(angle =45, hjust =1, vjust =1))
mean_sdl: Calculates the mean plus or minus two standard deviations (in this case, for each region)
Mean vaccination rates were highest in the New England and New York Islands, while they were lowest in the Southeast.
covid_state %>%ggplot() +stat_summary(aes(x = region_name, y = deaths),fun.data ="mean_sdl",geom ="errorbar") +labs(x ="Region Name", y ="Mean Death Rate",title ="Mean Death Rate, Per Region") +theme(axis.text.x =element_text(angle =45, hjust =1, vjust =1))
Mean average death rates were highest in the South Central and Southeast region. The South Central and Central Plains had the lowest standard deviations from the mean average death rates.
7. Make a map showing the spatial trend in COVID deaths in the US
Use geom_polygon and map_data. Modify the given code.
Merge the map_data data.frame with your COVID data.frame to add in the COVID data to the map data. Hint: use left_join.
Use a color palette with custom colors
Make sure that your legend is labelled correctly.
Add a * or other label for the top 10 states in highest death rate. !!!!!!!!!!!!
Make sure that the aspect ratio is appropriate
Describe your observations
Describe the trend in COVID death rates across the US
library(maps)
Attaching package: 'maps'
The following object is masked from 'package:purrr':
map
library(stringr)US_map <-map_data("state")US_map$region <-str_to_title(US_map$region)US_map <-left_join(US_map, covid_state, by =c("region"="state_name")) %>%mutate(state = region)top_10 <- US_map %>%summarize(mean_long =mean(long),mean_lat =mean(lat),.by =c("state", "region_name", "deaths")) %>%arrange(deaths) %>%head(10)US_map %>%filter(!is.na(region_name)) %>%ggplot(aes(x = long, y = lat, group = group, fill = region_name)) +scale_fill_brewer(name ="Region Name", palette ="PiYG") +geom_polygon(color ="white") +geom_text(data = top_10,aes(x = mean_long, y = mean_lat,label ="*"),inherit.aes =FALSE,color ="black", size =6) +theme(aspect.ratio =0.75)
8. Use a ggplot extension
Pick an extension (except cowplot) from here and make a plot of your choice using the COVID data
You might want to try examples that come with the extension first (e.g. ggtech, gganimate, ggforce)
